Tlie Spermatogenesis of the Opossum (Didelphys virginiana) etc. 75 
my observations convince me tliat the originally separate grannles become 
fused into a continuous filament 10 ) X1 ). 
It appears that, despite great morpliological variations (and certain 
obscurities respecting details), the spermiohistogenetic process is very 
similar in many mammals; i. e. the archoplasmic remnant (centrosphere) 
of the second maturation mitosis separates into idiozome (frorn which 
separate the centrosomes: centrioles) and Xebenkern. Tire latter becomes 
the acrosome (a portion perhaps contributing to the nriddle-piece), the 
former seems to degenerate and disappear. Duesberg describes and 
illustrates also certain “chromatoid bodies” in the secondary spermato- 
cytes and spernratids of the rat and the guinea-pig, and interprets them 
as the persisting nucleoh of the ”auxocytes”. Similar bodies were not 
definitely recognizable in my material of opossum testes. Leplat also 
was unable to identify such structirres in the testes of the cat. Xor can 
I differentiate from mitochondria the “tingierbare Körner” of von Ebner 
described by Duesberg for the rat and guinea-pig in the discarded “cyto- 
plasmic vesicle”. I believe that there is no question of such bodies in 
the opossum testes. The granules and spherules present are more pro- 
bablv chondriosomes. 
It is precisely here that Duesberg’s theory of the continuity of 
mitochondria, based upon a specifity of staining reaction to Benda’s 
stain, fails. In the degenerating mass of cytoplasm (cytoplasmic vesicle) 
are numerous bodies morphologically and tinctorially identical with the 
mitochondria of the middle-piece. Apparently, for Duesberg, those 
particles which pass into the middle-piece are mitochondria, not those 
which stain violet in Benda’s stain. Again, the so-called fat granules are 
morphologically similar to the mitochondria. If Benda’s stain colors 
the idiozome (Xebenkern — forms acrosome in guinea-pig) violet some- 
times and brown sometimes (the latter reaction is accepted as the proper 
oue, hence the idiozome is not here regarded as mitochondrial in nature) 
how can Duesberg be certain that the bodies like fat granules (similar 
to true chondriosomes morphologically) and the “tingierbare Körner” are 
not unstained mitochondria, or that the violet mitochondria-like bodies 
in the degenerating portion of the spermatid are not mitochondria? In 
the invertebrates the Xebenkern contributes to the middle-piece, hence 
Duesberg, after staining it violet, calls it mitochondrial. The Inter- 
pretation seems perfectly justified here since the X ebenkern can be traced 
back in the developmental history to independent chondriosomes. The 
point is that the staining reaction alone is not final proof of their true 
nature. On the other hand, in the guinea-pig where the Xebenkern (“idio- 
